{"title":"Optimizing Fuel Injection Timing for Multiple Injection Using\n Reinforcement Learning and Functional Mock-up Unit for a Small-bore Diesel\n Engine","authors":"Abhijeet Vaze, Pramod S. Mehta, Anand Krishnasamy","doi":"10.4271/03-17-06-0041","DOIUrl":null,"url":null,"abstract":"Reinforcement learning (RL) is a computational approach to understanding and\n automating goal-directed learning and decision-making. The difference from other\n computational approaches is the emphasis on learning by an agent from direct\n interaction with its environment to achieve long-term goals [1]. In this work, the RL algorithm was\n implemented using Python. This then enables the RL algorithm to make decisions\n to optimize the output from the system and provide real-time adaptation to\n changes and their retention for future usage. A diesel engine is a complex\n system where a RL algorithm can address the NOx–soot emissions\n trade-off by controlling fuel injection quantity and timing. This study used RL\n to optimize the fuel injection timing to get a better NO–soot trade-off for a\n common rail diesel engine. The diesel engine utilizes a pilot–main and a\n pilot–main–post-fuel injection strategy. Change of fuel injection quantity was\n not attempted in this study as the main objective was to demonstrate the use of\n RL algorithms while maintaining a constant indicated mean effective pressure. A\n change in fuel quantity has a larger influence on the indicated mean effective\n pressure than a change in fuel injection timing. The focus of this work was to\n present a novel methodology of using the 3D combustion data from analysis\n software in the form of a functional mock-up unit (FMU) and showcasing the\n implementation of a RL algorithm in Python language to interact with the FMU to\n reduce the NO and soot emissions by suggesting changes to the main injection\n timing in a pilot–main and pilot–main–post-injection strategy. RL algorithms\n identified the operating injection strategy, i.e., main injection timing for a\n pilot–main and pilot–main–post-injection strategy, reducing NO emissions from\n 38% to 56% and soot emissions from 10% to 90% for a range of fuel injection\n strategies.","PeriodicalId":47948,"journal":{"name":"SAE International Journal of Engines","volume":null,"pages":null},"PeriodicalIF":1.1000,"publicationDate":"2024-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SAE International Journal of Engines","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4271/03-17-06-0041","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"TRANSPORTATION SCIENCE & TECHNOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Reinforcement learning (RL) is a computational approach to understanding and
automating goal-directed learning and decision-making. The difference from other
computational approaches is the emphasis on learning by an agent from direct
interaction with its environment to achieve long-term goals [1]. In this work, the RL algorithm was
implemented using Python. This then enables the RL algorithm to make decisions
to optimize the output from the system and provide real-time adaptation to
changes and their retention for future usage. A diesel engine is a complex
system where a RL algorithm can address the NOx–soot emissions
trade-off by controlling fuel injection quantity and timing. This study used RL
to optimize the fuel injection timing to get a better NO–soot trade-off for a
common rail diesel engine. The diesel engine utilizes a pilot–main and a
pilot–main–post-fuel injection strategy. Change of fuel injection quantity was
not attempted in this study as the main objective was to demonstrate the use of
RL algorithms while maintaining a constant indicated mean effective pressure. A
change in fuel quantity has a larger influence on the indicated mean effective
pressure than a change in fuel injection timing. The focus of this work was to
present a novel methodology of using the 3D combustion data from analysis
software in the form of a functional mock-up unit (FMU) and showcasing the
implementation of a RL algorithm in Python language to interact with the FMU to
reduce the NO and soot emissions by suggesting changes to the main injection
timing in a pilot–main and pilot–main–post-injection strategy. RL algorithms
identified the operating injection strategy, i.e., main injection timing for a
pilot–main and pilot–main–post-injection strategy, reducing NO emissions from
38% to 56% and soot emissions from 10% to 90% for a range of fuel injection
strategies.